INDEX

Explanations

references to monsters and horror elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

monsters and terrifying creatures

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sidemargin

-0.56

IBOutlet

-0.44

RenderAtEndOf

-0.44

хьтан

-0.43

 exemples

-0.43

 GenerationType

-0.43

balleur

-0.43

tangentMode

-0.41

 Artículos

-0.41

 muhimu

-0.40

POSITIVE LOGITS

 monsters

0.63

 monster

0.58

 terrifying

0.57

 frightening

0.50

monster

0.50

 menacing

0.50

Datuak

0.49

 scary

0.48

 monstre

0.48

 horrifying

0.46

Activations Density 0.051%

references to monsters and horror elements

monsters and terrifying creatures

No Comments

No Known Activations